1. Field of the Invention
The present invention relates generally to a computer implemented method, data processing system, and computer program product for providing file system support to applications. More specifically, the present invention relates to collecting metadata.
2. Description of the Related Art
Modern uses of networked computers leverage storage capacity by sharing groups of storage devices together to better use available capacity of the collection of storage devices. Technologies such as storage area network (SAN) and network attached storage (NAS), for example, have driven demand for collectivizing storage devices into pools accessed over the network. Organized and managed by a small number of storage servers, many storage devices respond to requests made by applications that often reside on separate processors or hosts.
Application level programs such as database programs can rely heavily on pools of storage devices to provide storage as a service. Although this arrangement can allow an organization to achieve greater focus on tuning and maintaining the application, the arrangement can create impediments to efficiently eliminating waste caused by duplicating storage objects in the pool of storage devices. In addition, application level programs can be challenged to identify changed portions of a repository that may require backing up from time to time.
These challenges develop from a lack of data at the application level to identify, in an efficient manner, storage objects that are unique, and storage objects that are identical. Prior art has shown that metadata can be accessed by way of using a stat command in the AIX® operating system, among others. AIX is a trademark of the IBM Corporation in the United States, other countries, or both. A stream is a sequence of data of undetermined length. An attribute stream is a sequence of metadata or data about data.